{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,28]],"date-time":"2026-02-28T19:23:27Z","timestamp":1772306607190,"version":"3.50.1"},"reference-count":41,"publisher":"MDPI AG","issue":"6","license":[{"start":{"date-parts":[[2023,5,25]],"date-time":"2023-05-25T00:00:00Z","timestamp":1684972800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.mdpi.com"],"crossmark-restriction":true},"short-container-title":["J. Imaging"],"abstract":"<jats:p>Ancient numismatics, the study of ancient coins, has in recent years become an attractive domain for the application of computer vision and machine learning. Though rich in research problems, the predominant focus in this area to date has been on the task of attributing a coin from an image, that is of identifying its issue. This may be considered the cardinal problem in the field and it continues to challenge automatic methods. In the present paper, we address a number of limitations of previous work. Firstly, the existing methods approach the problem as a classification task. As such, they are unable to deal with classes with no or few exemplars (which would be most, given over 50,000 issues of Roman Imperial coins alone), and require retraining when exemplars of a new class become available. Hence, rather than seeking to learn a representation that distinguishes a particular class from all the others, herein we seek a representation that is overall best at distinguishing classes from one another, thus relinquishing the demand for exemplars of any specific class. This leads to our adoption of the paradigm of pairwise coin matching by issue, rather than the usual classification paradigm, and the specific solution we propose in the form of a Siamese neural network. Furthermore, while adopting deep learning, motivated by its successes in the field and its unchallenged superiority over classical computer vision approaches, we also seek to leverage the advantages that transformers have over the previously employed convolutional neural networks, and in particular their non-local attention mechanisms, which ought to be particularly useful in ancient coin analysis by associating semantically but not visually related distal elements of a coin\u2019s design. Evaluated on a large data corpus of 14,820 images and 7605 issues, using transfer learning and only a small training set of 542 images of 24 issues, our Double Siamese ViT model is shown to surpass the state of the art by a large margin, achieving an overall accuracy of 81%. Moreover, our further investigation of the results shows that the majority of the method\u2019s errors are unrelated to the intrinsic aspects of the algorithm itself, but are rather a consequence of unclean data, which is a problem that can be easily addressed in practice by simple pre-processing and quality checking.<\/jats:p>","DOI":"10.3390\/jimaging9060107","type":"journal-article","created":{"date-parts":[[2023,5,25]],"date-time":"2023-05-25T01:36:28Z","timestamp":1684978588000},"page":"107","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":7,"title":["A Siamese Transformer Network for Zero-Shot Ancient Coin Classification"],"prefix":"10.3390","volume":"9","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6025-3021","authenticated-orcid":false,"given":"Zhongliang","family":"Guo","sequence":"first","affiliation":[{"name":"School of Computer Science, University of St Andrews, Scotland KY16 9AJ, UK"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9314-194X","authenticated-orcid":false,"given":"Ognjen","family":"Arandjelovi\u0107","sequence":"additional","affiliation":[{"name":"School of Computer Science, University of St Andrews, Scotland KY16 9AJ, UK"}]},{"given":"David","family":"Reid","sequence":"additional","affiliation":[{"name":"School of Computer Science, University of St Andrews, Scotland KY16 9AJ, UK"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0697-7942","authenticated-orcid":false,"given":"Yaxiong","family":"Lei","sequence":"additional","affiliation":[{"name":"School of Computer Science, University of St Andrews, Scotland KY16 9AJ, UK"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1758-3153","authenticated-orcid":false,"given":"Jochen","family":"B\u00fcttner","sequence":"additional","affiliation":[{"name":"Max Planck Institute for the History of Science, Boltzmannstra\u00dfe 22, 14195 Berlin, Germany"}]}],"member":"1968","published-online":{"date-parts":[[2023,5,25]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Arandjelovi\u0107, O., and Zachariou, M. (2020). Images of Roman imperial denarii: A curated data set for the evaluation of computer vision algorithms applied to ancient numismatics, and an overview of challenges in the field. Sci, 2.","DOI":"10.3390\/sci2040091"},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Huber-M\u00f6rk, R., N\u00f6lle, M., Rubik, M., H\u00f6dlmoser, M., Kampel, M., and Zambanini, S. (2012). Automatic coin classification and identification. Advances in Object Recognition Systems, Oxford University Press.","DOI":"10.5772\/35795"},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Kiourt, C., and Evangelidis, V. (2021). AnCoins: Image-Based Automated Identification of Ancient Coins Through Transfer Learning Approaches. Pattern Recognition, Proceedings of the ICPR International Workshops and Challenges, Virtual Event, 10\u201315 January 2021, Springer.","DOI":"10.1007\/978-3-030-68787-8_4"},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Wei, K., He, B., Wang, F., Zhang, T., and Ding, Q. (2007, January 10\u201312). A novel method for classification of ancient coins based on image textures. Proceedings of the Workshop on Digital Media and Its Application in Museum & Heritages, Chongqing, China.","DOI":"10.1109\/DMAMH.2007.4414528"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Zaharieva, M., Kampel, M., and Zambanini, S. (2007). Image based recognition of ancient coins. Computer Analysis of Images and Patterns, Proceedings of the 12th International Conference, CAIP 2007, Vienna, Austria, 27\u201329 August 2007, Springer.","DOI":"10.1007\/978-3-540-74272-2_68"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Arandjelovi\u0107, O. (2012). Reading ancient coins: Automatically identifying denarii using obverse legend seeded retrieval. Computer Vision\u2013ECCV 2012, Proceedings of the 12th European Conference on Computer Vision, Florence, Italy, 7\u201313 October 2012, Springer.","DOI":"10.1007\/978-3-642-33765-9_23"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Cooper, J., and Arandjelovi\u0107, O. (2020). Understanding ancient coin images. Recent Advances in Big Data and Deep Learning, Proceedings of the INNS Big Data and Deep Learning Conference INNSBDDL2019, held at Sestri Levante, Genova, Italy 16\u201318 April 2019, Springer.","DOI":"10.1007\/978-3-030-16841-4_34"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Schlag, I., and Arandjelovic, O. (2017, January 22\u201329). Ancient Roman coin recognition in the wild using deep learning based recognition of artistically depicted face profiles. Proceedings of the International Conference on Computer Vision Workshops, Venice, Italy.","DOI":"10.1109\/ICCVW.2017.342"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"158","DOI":"10.1016\/j.patrec.2019.12.007","article-title":"Two sides of the same coin: Improved ancient coin classification using Graph Transduction Games","volume":"131","author":"Aslan","year":"2020","journal-title":"Pattern Recognit. Lett."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Cooper, J., and Arandjelovi\u0107, O. (2020). Learning to Describe: A New Approach to Computer Vision Based Ancient Coin Analysis. Sci, 2.","DOI":"10.3390\/sci2020027"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Kampel, M., and Zaharieva, M. (2008). Recognizing ancient coins based on local features. Advances in Visual Computing, Proceedings of the 4th International Symposium, ISVC 2008, Las Vegas, NV, USA, 1\u20133 December 2008, Springer.","DOI":"10.1007\/978-3-540-89639-5_2"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"26","DOI":"10.1109\/MIS.2009.29","article-title":"Image-based retrieval and identification of ancient coins","volume":"24","author":"Kampel","year":"2009","journal-title":"IEEE Intell. Syst."},{"key":"ref_13","unstructured":"Zambanini, S., and Kampel, M. (2009, January 5\u20138). Robust Automatic Segmentation of Ancient Coins. Proceedings of the International Conference on Computer Vision Theory and Applications, Lisboa, Portugal."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"983","DOI":"10.1007\/s00138-010-0283-y","article-title":"Identification of ancient coins based on fusion of shape and local features","volume":"22","author":"Zambanini","year":"2011","journal-title":"Mach. Vis. Appl."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Anwar, H., Zambanini, S., and Kampel, M. (2013). Supporting ancient coin classification by image-based reverse side symbol recognition. Computer Analysis of Images and Patterns, Proceedings of the 15th International Conference, CAIP 2013, York, UK, 27\u201329 August 2013, Springer.","DOI":"10.1007\/978-3-642-40246-3_3"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Zachariou, M., Dimitriou, N., and Arandjelovi\u0107, O. (2020). Visual reconstruction of ancient coins using cycle-consistent generative adversarial networks. Sci, 2.","DOI":"10.3390\/sci2030052"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"64","DOI":"10.1109\/MSP.2015.2409331","article-title":"Ancient coin classification using reverse motif recognition: Image-based classification of roman republican coins","volume":"32","author":"Anwar","year":"2015","journal-title":"IEEE Signal Process. Mag."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Conn, B., and Arandjelovi\u0107, O. (2017, January 14\u201319). Towards computer vision based ancient coin recognition in the wild\u2014Automatic reliable image preprocessing and normalization. Proceedings of the International Joint Conference on Neural Networks, Anchorage, AK, USA.","DOI":"10.1109\/IJCNN.2017.7966024"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Arandjelovi\u0107, O. (2010, January 13\u201318). Automatic attribution of ancient Roman imperial coins. Proceedings of the Computer Vision and Pattern Recognition Conference, San Francisco, CA, USA.","DOI":"10.1109\/CVPR.2010.5539841"},{"key":"ref_20","unstructured":"Zaharieva, M., Huber-M\u00f6rk, R., N\u00f6lle, M., and Kampel, M. (2007, January 26\u201330). On ancient coin classification. Proceedings of the International Symposium on Virtual Reality, Archaeology and Intelligent Cultural Heritage, Brighton, UK."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Anwar, H., Zambanini, S., and Kampel, M. (2014). Encoding spatial arrangements of visual words for rotation-invariant image classification. Pattern Recognition, Proceedings of the 36th German Conference, GCPR 2014, M\u00fcnster, Germany, 2\u20135 September 2014, Springer.","DOI":"10.1007\/978-3-319-11752-2_36"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Ma, Y., and Arandjelovi\u0107, O. (2020). Classification of ancient roman coins by denomination using colour, a forgotten feature in automatic ancient coin analysis. Sci, 2.","DOI":"10.3390\/sci2020037"},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Fare, C., and Arandjelovi\u0107, O. (2017). Ancient roman coin retrieval: A systematic examination of the effects of coin grade. Advances in Information Retrieval, Proceedings of the 39th European Conference on IR Research, ECIR 2017, Aberdeen, UK, 8\u201313April 2017, Springer.","DOI":"10.1007\/978-3-319-56608-5_32"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"011018","DOI":"10.1117\/1.JEI.26.1.011018","article-title":"Discovering characteristic landmarks on ancient coins using convolutional networks","volume":"26","author":"Kim","year":"2017","journal-title":"J. Electron. Imaging"},{"key":"ref_25","first-page":"737","article-title":"Signature verification using a \u201cSiamese\u201d time delay neural network","volume":"6","author":"Bromley","year":"1993","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_26","first-page":"6000","article-title":"Transformer: Attention is all you need","volume":"30","author":"Vaswani","year":"2017","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Chicco, D. (2021). Siamese neural networks: An overview. Artificial Neural Networks, Springer.","DOI":"10.1007\/978-1-0716-0826-5_3"},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Zhang, C., Liu, W., Ma, H., and Fu, H. (2016, January 20\u201325). Siamese neural network based gait recognition for human identification. Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing, Shanghai, China.","DOI":"10.1109\/ICASSP.2016.7472194"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"3649","DOI":"10.1007\/s12652-019-01575-w","article-title":"Research on application of athlete gesture tracking algorithms based on deep learning","volume":"11","author":"Long","year":"2020","journal-title":"J. Ambient. Intell. Humaniz. Comput."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Ichida, A.Y., Meneguzzi, F., and Ruiz, D.D. (2018, January 8\u201313). Measuring semantic similarity between sentences using a Siamese neural network. Proceedings of the International Joint Conference on Neural Networks, Rio de Janeiro, Brazil.","DOI":"10.1109\/IJCNN.2018.8489433"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"336","DOI":"10.1016\/j.patrec.2020.01.012","article-title":"Matching ostraca fragments using a Siamese neural network","volume":"131","author":"Ostertag","year":"2020","journal-title":"Pattern Recognit. Lett."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Berlemont, S., Lefebvre, G., Duffner, S., and Garcia, C. (2015, January 4\u20138). Siamese neural network based similarity metric for inertial gesture classification and rejection. Proceedings of the IEEE International Conference and Workshops on Automatic Face and Gesture Recognition, Ljubljana, Slovenia.","DOI":"10.1109\/FG.2015.7163112"},{"key":"ref_33","unstructured":"Kim, M., Alletto, S., and Rigazio, L. (2016). Similarity mapping with enhanced Siamese network for multi-object tracking. arXiv."},{"key":"ref_34","unstructured":"Devlin, J., Chang, M.W., Lee, K., and Toutanova, K. (2018). Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv."},{"key":"ref_35","first-page":"1877","article-title":"Language models are few-shot learners","volume":"33","author":"Brown","year":"2020","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_36","first-page":"6391","article-title":"Visualizing the loss landscape of neural nets","volume":"31","author":"Li","year":"2018","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_37","unstructured":"Ba, J.L., Kiros, J.R., and Hinton, G.E. (2016). Layer normalization. arXiv."},{"key":"ref_38","unstructured":"Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., and Gelly, S. (2020). An image is worth 16x16 words: Transformers for image recognition at scale. arXiv."},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., and Sun, J. (2016, January 27\u201330). Deep residual learning for image recognition. Proceedings of the Computer Vision and Pattern Recognition Conference, Las Vegas, NV, USA.","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"388","DOI":"10.1016\/j.patcog.2018.06.006","article-title":"Reimagining the central challenge of face recognition: Turning a problem into an advantage","volume":"83","year":"2018","journal-title":"Pattern Recognit."},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Arandjelovic, O. (2016, January 27\u201330). Learnt quasi-transitive similarity for retrieval from large collections of faces. Proceedings of the Computer Vision and Pattern Recognition Conference, Las Vegas, NV, USA.","DOI":"10.1109\/CVPR.2016.528"}],"updated-by":[{"DOI":"10.3390\/jimaging10030057","type":"correction","label":"Correction","source":"publisher","updated":{"date-parts":[[2023,5,25]],"date-time":"2023-05-25T00:00:00Z","timestamp":1684972800000}}],"container-title":["Journal of Imaging"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2313-433X\/9\/6\/107\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,3]],"date-time":"2025-08-03T14:36:11Z","timestamp":1754231771000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2313-433X\/9\/6\/107"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,5,25]]},"references-count":41,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2023,6]]}},"alternative-id":["jimaging9060107"],"URL":"https:\/\/doi.org\/10.3390\/jimaging9060107","relation":{"correction":[{"id-type":"doi","id":"10.3390\/jimaging10030057","asserted-by":"object"}]},"ISSN":["2313-433X"],"issn-type":[{"value":"2313-433X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,5,25]]}}}